FIG. 1 shows a basic mobile internet protocol (IP) system. Mobile IP version 6 (referred to as MIPv6 hereinafter) is a method for processing IP node mobility that is defined by the Internet Engineering Task Force (IETF) [“RFC3775: Mobility Support in IPv6”, June, 2004, IETF]. All mobile nodes (MN) 11 using the mobile IP (MIP) protocol have corresponding home agents (HA) 12 and home IP addresses (HoA), wherein, the route of the HoA address in the network topology points to the HA.
IP sub-network which the MN accesses becomes a “visited network (VN)”. The HA corresponding to the MN is located outside any VN that may be accessed. When a MN is located within a certain VN, the MN cannot use its HoA as a valid IP address, because the route of the HoA in the network topology does not point to the VN. Therefore, the MN configures a care-of address (CoA) using a valid local prefix. As long as the MN resides within the same visited IP network, the CoA can be used to transmit and receive an IP message. Change of the VN will cause IP switching. “Correspondent node (CN)” refers to an IP host that is communicating with the above MN. One MN can communicate with a plurality of CNs simultaneously, and the MN per se is movable.
The MN using MIP can select among two communication modes, so as to keep the session continuity in the case of IP mobility. In the first case, all services between the CN and MN are forwarded by the HA using a bidirectional channel. Every time the CoA of the MN changes, the MN updates the HoA-CoA mapping relationship in the HA using a binding update (BU) message. The HA forwards all received IP messages with the HoA being the destination address to the MN using the actual CoA channel of the MN. Similarly, the MN uses the HoA as the internal source IP address, and sends all sent-out messages to the HA through the channel. The HA de-encapsulates the data message and sends the packet to its final destination IP address. If the HA is trusted by the MN, this communication mode can realize location privacy. However, a disadvantage of this communication mode is suboptimality of the routing: all data messages need to pass through the home agent HA, and thus there is routing alternate (or called as triangle routing).
The second communication mode in the MIP is routing optimization. Routing optimization allows the data message to be directly transmitted and received between the MN and the CN without visit to the HA. Routing optimization is implemented by sending a BU message from the MN to the CN. The BU message contains the CoA address of the MN. Afterwards, the data message between the MN and the CN adopts the CoA address as a source address or a destination address, and such data message is sent using an optimal routing. Since the location of the MN (CoA, for indicating the current location of the MN in the network topology) is disclosed to the CN, this process violates location privacy.
The above-mentioned MIP system of the IETF standard does not allow combination of location privacy and routing optimization.
Meanwhile, the mobile control system using “reverse channel” mobile IP protocol has a problem, namely: the source mobile terminal (MT) #1 has to reconfigure the care-of address of the mobile node MN and the care-of address of the destination mobile terminal MT#2 in the data message header, so the data message header is very big.
In order to solve the above problem, the patent application titled “A method and apparatus for route grouping in mobile IP system” filed on Aug. 9, 2006, with a publication No. 101507226 and a filing No. of 200680055577.6 proposes a solution.
As shown in FIG. 2, by adding functional nodes such as NCPF (RM) 21 and IP2P, the IP2 system structure and MIP/layered MIP (HMIP) system are combined in this solution to provide a method, device and system for optimally solving this problem. In the invention, IPha is a unique permanent host identifier assigned to the MN, and IPra is a routing host address. The routing optimization flow initiated by MN is monitored through a monitoring unit provided inside the IP2P. When the MN initiates the routing optimization, the IP2P detects and intercepts the BU message, and sends an activation notification to the NCPF (RM), the NCPF (RM) sends a routing optimization instruction to the related IP2P after receiving the activation notification, and the related IP2P optimizes the forwarding routing of the data message based on the received routing optimization instruction. As a result, the data message is implemented on MN→IP2P_MN→IP2P_CN→CN. Therefore, the routing path via the IP2P is shorter than the forwarding path via the home agent HA. The method for forwarding a data message between IP2P may be address exchange, tunnel delivery or secure tunnel delivery. In addition, the routing addresses (CoA, LCoA, etc.) of the MN are applied safely, so the location privacy of the MN is preferably protected. The main problems of this solution are as follows:
1-1 All data messages sent out by the MN need to be traversed in order to monitor the routing optimization flow initiated by the MN, which will cost a large amount of processing performance of the IP2P.
1-2 the communication terminal on which the invention is based must be a MIP terminal, or a communication terminal with a particular functional requirement based on a similar technique principle, and the implementation process needs participation of the terminal.
1-3 The routing optimization of MIP is an end-to-end process, which requires the end to end to support the MIP routing optimization protocol, and there is still routing alternate when the MIP terminal does not perform routing optimization or does not possess the routing optimization condition (for example, the correspondent node is a fixed access terminal that does not support the MIP protocol, or the version of the MIP protocol supported by the correspondent node is different).
A patent application for invention titled “Mobile communication control system and method, routing device, server and data structure” with a filing No. of 03157408.4 and publication No. of 1496154 was filed on Sep. 19, 2003.
In the patent solution, in considering the problems existing in the mobile IP technology including route redundancy and leak of the current location information of the mobile node (mobile terminal) and the IP address of the local proxy server, the following steps are defined: converting a first address of a destination mobile terminal 2 included in a packet as a destination address, to a second address of the destination mobile terminal 2 which is associated with the first address of the destination mobile terminal 2, in a source router (AR2); and converting the second address of the destination mobile terminal 2 included in the received packet as a destination address, to the first address of the destination mobile terminal 2 which is associated with the second address of the destination mobile terminal 2, in a router AR3.
This technical scheme has the following defects:
2-1 This scheme provides a method for converting between the first address and the second address, and the judging bits in the IPv6 address are redefined to distinguish the first address from the second address. The accepted meaning of the 128-bit address of the IPv6 has been explicitly defined in the IETF standard organization, wherein the judging bits are defined, which on one hand decreases the number of available IPv6 addresses, but on the other hand arises the possibility of being conflicting to the existing protocol specifications, so the IPv6 address definition in the existing regulation standards needs to be modified.
2-2 The network management server NCPF maintains a line selecting table managing unit, and sends a notification message to a related router through a router device control unit to optimize the routing transmission path. During the process of routing optimization and routing information maintenance, update and deletion, the network management server NCPF will interact with the related router, but how the network management server NCPF obtains the related router is not specifically introduced. The network management server NCPF can only obtain the related router information by interacting with the terminal or the router AR which the terminal accesses, which demands a complicated signaling interaction process.
2-3 The process of optimizing the routing transmission path is triggered by one of the access routers of the two communication terminals or one of the communication terminals. The network management server NCPF sends out a notification message for optimizing the routing transmission path, and then the access router of the two communication terminals executes the operation of optimizing the routing according to the notification message. Wherein participation of the network server NCPF in this process is not necessary, and participation of the network server NCPF causes unnecessary fault nodes, and makes this process more complicated.
2-4 Each terminal will have both the IP host address and IP routing address, and thus occupies two IP addresses, which causes certain resource waste.
2-5 When the IP host address is not stored in the buffer table of the transmission storage unit, the address judging unit stores the IP packet in a buffer temporarily. At this moment, the address judging unit queries the location information management server to obtain the IP routing address (the second address) corresponding to the IP host address as the destination address. According to the query result, the destination address in the IP packet is converted from the IP host address into the IP routing address, and is buffered in the buffer, and the IP packet is sent to the routing unit.
The number of IP packets stored in the buffer is closely related with the time of query and the rate of sending the messages by the mobile terminal. Since these stored IP packets need to occupy the performance of the storage unit and also need complicated processing such as queue management, big processing load and complexity will be brought to the device with the background of increasing wireless bandwidth and wired access bandwidth. Meanwhile the storing and forwarding mode causes increase of the forwarding delay of the data message.
The patent application for invention titled “Mobile communication control system, network management server and node” field on Feb. 4, 2004, with a filing No. of 200410003107.8 and publication No. of 1523837 is a specific scene of the above patent application for invention of “Mobile communication control system and method, routing device, server and data structure”, its basic principle is the same as the above patent application solution, and it has the features 2-2, 2-3 and 2-4 of the above patent application.
The patent application titled “Method and apparatus for route grouping in mobile IP system” filed on Aug. 9, 2006 with a publication No. of 101513006 and a filing No. of 200680055560.0 proposes a method for solving the problem in the switching scene in the IP2 solution. It has the following features:
3-1 The basic method for converting the address of the data message is using IPra to replace IPha.
3-2 The routing optimization process is implemented as follows: after nAR113 initiates an activation notification AN to the RM121 through step S202, RM121 sends the IP update in step S204 to the opposite end AR111. RM121 is the main body for initiating the routing optimization process.
3-3 How RM121 obtains the information of the opposite end AR111 and to which AR111 the routing optimization process is initiated are not explicitly described in the patent solution.
It has the defects 2-2, 2-3 and 2-4 of the above patent application solution.